LGI does not protect against use-after-free issues that can occur due to
using an object after finalisation. This manifests itself as occasional
crashes on Travis in cairo_region_union_rectangle() (AFAIK no one ran
into this issue in real-world usage).
Since visible drawables are always strongly reachable, the issue can
only occur with invisible drawables. The previous commit made sure that
those are fully repainted when they become visible, so we can just
ignore redraws for those and fix the crash issue.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Instead of tracking all drawables that are alive, the code now only
tracks visible drawables. When a drawable is made visible it is
completely repainted. This should not cause a difference when a wibox is
initially made visible, because it has to be redrawn anyway. However,
this introduces a full repaint when a wibox is hidden and then made
visible again.
Thanks to this change, we can stop using weak tables. Visible drawables
cannot be collected and so we can keep a strong reference to them. This
allows us to get rid of the weak tables which solves various problems
involving finalizers and using objects after finalisation.
Signed-off-by: Uli Schlachter <psychon@znc.in>
This new function is called whenever the visibility of the drawable
changes. Later commits can use this for explicitly tracking the lifetime
of drawables instead of using magic weak tables.
Signed-off-by: Uli Schlachter <psychon@znc.in>
This is the first step in deprecating them. A function with so
many optional arguments is just bad design.
The next few commits will rewrite the documentation and deprecate
the old arguments.
For a while, it was often suggested on IRC to replace the default
request::activate handler to implement custom focus stealing policies.
While it is working, it isn't user friendly. This commit add a simple
mechanism to add such policies.
This adds a tparam alias "@screen" for "@tparam screen" (when used to
document e.g. arguments for callbacks), and "@screen_or_idx" when a
function accepts a "screen" or "number".
The default config had tables like mywibox and mywibox[s] was the wibox
that is visible on screen s. When a screen is removed, nothing cleans up
these tables and so the screen and the wibox could not be garbage
collected. The same applies to the layoutbox, taglist etc.
This commit removes the global mywibox table and instead saves it as a
property on the screen. This way, the screen is not explicitly
referenced and when it is removed, the screen, its wibox and all of its
widgets become unreachable and can be garbage collected.
This commit also updates the docs and the tests that referenced things
(mostly the wibox) via mywibox[s] to now use s.mywibox.
Fixes: https://github.com/awesomeWM/awesome/issues/1125
Signed-off-by: Uli Schlachter <psychon@znc.in>
Previously, gears.object.properties used a weak table for adding
additional information to a C object. However, weak tables can easily
cause leaks when the value references the key.
This commit makes the code instead use the new .data property that is
available on all C objects. This means we have no more magic with a weak
table and instead only use "regular" tables instead.
Signed-off-by: Uli Schlachter <psychon@znc.in>
This restore a feature that was available in Awesome 2.1-3.2.
The reason margin is implemented rather than use a container is to
be able to make the background smaller than the bar.
The current progressbar code dates from a time when Awesome had
a very limited drawing API. This commit first re-write the
algorithm to remove the workaround used to draw the border using
full rectangles only. It then add support for outer and inner
shapes with their respective border settings.
This commit also add clip support. This is enabled by default, but
could be disabled to have the bar taller than the background.
Twice now we had problems with the garbage collector which caused signals
established via weak_connect_signal() not to be disconnected when we wanted them
to be disconnected. The effect was that we tried to redraw a drawable after it
was garbage collected which caused errors.
Instead of playing whack-a-mole with all the various ways that might make us
redraw a drawable after GC, let's just fix all of these issues by explicitly
checking for this case and turning it into a no-op.
Signed-off-by: Uli Schlachter <psychon@znc.in>
The previous commit made wibox.drawable turn a "normal redraw" into a complete
repaint when it was moved to another screen. However, nothing happened until
that normal redraw.
This commit triggers a normal redraw when we are (possibly) moved to another
screen. More precise, this means that whenever a screen appears, disappears or
changes its geometry and when the drawable is moved, we trigger a normal redraw.
This redraw will likely do nothing, because no relayout is pending and no part
of the surface needs a redraw, so it is cheap.
However, if the drawable really ends up on another screen, then the code from
the previous commits makes us do a full relayout and redraw.
This commit likely fixes the current instability of test-screen-changes.lua. See
https://github.com/awesomeWM/awesome/issues/982#issuecomment-231712056.
As explained there, the test fails because the fake screen that it created is
still referenced, so cannot be garbage collected, but the test doesn't succeed
unless the screen is garbage collected. So something is still referencing the
screen that was removed. This something can be a client's titlebar, because the
underlying drawable still has a context member referring to the old screen.
This commit should fix that problem, because we now trigger a redraw which will
compute a new context and thus the reference to the old screen is released.
Signed-off-by: Uli Schlachter <psychon@znc.in>
The previous commit made the hierarchy do a re-layout when the context changes.
However, widgets could change their appearance depending on the context without
changing their layout. Thus, the previous commit is not enough.
This commit also makes the drawable redraw everything when the context changes.
Signed-off-by: Uli Schlachter <psychon@znc.in>
When the context for widget changes (e.g. we are on a new different screen or
have a different DPI value), widgets might change their appearance even though
they didn't emit widget::layout_changed. Thus, update the hierarchy in these
cases.
Signed-off-by: Uli Schlachter <psychon@znc.in>
widget_at() no longer exists since 0aa4304bda (and the surrounding commits
stopped us using this function).
Signed-off-by: Uli Schlachter <psychon@znc.in>
The magnifier layout wants to ignore floating clients. Before 82342f0 this was
done by calling awful.client.floating.get(focus). If "focus" was nil, this might
have checked the floating status of a wrong client (if some other client was
focused, and the code in magnifier set focus=nil before). This issue can easily
be missed and might exist since forever. After 82342f, floating status is
checked via "focus.floating" and this now causes an "attempt to index nil value"
error instead. Much easier to notice.
Fix this by adding the missing nil check and while touching the code, merge this
with the previous "if" and correct another error (the wrong thing happened if we
had #cls=0).
Fixes: https://github.com/awesomeWM/awesome/issues/1103
Signed-off-by: Uli Schlachter <psychon@znc.in>
A client is supposed to go to a screen when:
* It has been started using `awful.spawn` with explicit instructions [1]
* An `awful.rules` rule **or any of its callbacks** set the screen [2]
* When something handle `request::screen` and/or `request::tag` in some
custom ways. [3]
* Some clients can request a screen and mean it (like MythTV/Kodi/XBMC and
some multi-window DAW) [4]
A client is supposed to go to the focused screen when none of the above are
true [5].
Other constraints:
* The screen need to be set only once, anything will will emit
`property::screen` many time and cause side effects.
* There has to be a single entry point to the algorithm, no multiple
"manage" handler.
* Awesome internals must use the `request::` signal API and not force
their decision outside of request handlers.
* Restarting Awesome must not change the client screen
Commit 2178744 fix use case number [1] and [2]. It actually fix [4] too, but
it is an accident and I am not sure we care about [4] anyway. Use case [1]
and [2], however, are very important.
Fix#1091
The geometry storage has been moved into awful.placement. This
code was never executed as data[] was never populated.
There is some behavior that is indeed lost, but it is unlikely
someone will ever notice (it has been broken for 6 months).
The previous code attempted to handle scrren changes while
maximized. The new code organization shift this responsability
to awful.placement. However, it doesn't yet fully implement the
previous logic.
Awesome 3.5.9 accepts `_active`/`_inactive` names for `beautiful`
minimize keys (such as "titlebar_minimize_button_focus_inactive").
Some themes rely on those, meaning that when they loaded under
the current Git, the minimize button went missing. This adds a
fallback, to improve compatibility with the existing themes.
This commit remove the `awful.tag` "manage" hook. The relevant
code has been moved to ewmh.lua request::tag handler. The handler
is called either by a volontary screen change or by a forced one.
It also require the awful.rules to be executed. This is done by
default and the user would have to explicitly disable that
behavior. From now on, disabling the rules require the user to
handle tag selection.
Fixes#1028#1052
There was still a problem that caused the "old" tags to be
inserted in the wrong position when "saved" from a screen being
removed.
Also, this use a :get_tags(true) to save an uneeded sorting pass.
The index was updated on an unordered table. As the elements
order did not match the relative indices once they have been
changed, further calls to set_index produced garbage.
The default taglist didn't notice because it use screen.tags
table index instead of the tag index. A debug using
echo 'for _,t in ipairs(mouse.screen.tags) do
print("INDEX:", _, t.index, t.name) end' | awesome-client
Would have shown two or more elements with the same index. To
debug issues related to tag indices, this bash script can be
enabled:
while true; do
echo 'for _,t in ipairs(mouse.screen.tags) do
assert( _==t.index) end' | awesome-client
sleep 0.5
done
Uli Schlachter
Emmanuel Lepage Vallee
Daniel Hahler
Daniel Hahler
Hisham
Akulen
modk
Akulen